Microcellular processing of polylactide–hyperbranchedpolyester–nanoclay composites

Srikanth Pilla • Adam Kramschuster •

Jungjoo Lee • Craig Clemons • Shaoqin Gong •

Lih-Sheng Turng

Received: 20 November 2009 / Accepted: 21 January 2010 / Published online: 4 February 2010

� Springer Science+Business Media, LLC 2010

Abstract The effects of addition of hyperbranched poly-

esters (HBPs) and nanoclay on the material properties of

both solid and microcellular polylactide (PLA) produced via

a conventional and microcellular injection-molding process,

respectively, were investigated. The effects of two different

types of HBPs (i.e., Boltorn H2004� and Boltorn H20�) at

the same loading level (i.e., 12%), and the same type of HBP

at different loading levels (i.e., Boltorn H2004� at 6 and

12%), as well as the simultaneous addition of 12% Boltorn

H2004� and 2% Cloisite�30B nanoclay (i.e., HBP–nano-

clay) on the thermal and mechanical properties (both static

and dynamic), and the cell morphology of the microcellular

components were noted. The addition of HBPs and/or HBP

with nanoclay decreased the average cell size, and increased

the cell density. The stress–strain plots of all the solid and

microcellular PLA-H2004 blends showed considerable

strain softening and cold drawing, indicating a ductile

fracture mode. Among the two HBPs, samples with Boltorn

H2004� showed higher strain-at-break and specific tough-

ness compared to Boltorn H20�. Moreover, the sample with

Boltorn H2004� and nanoclay exhibited the highest strain-

at-break (626% for solid and 406% for microcellular) and

specific toughness (405% for solid and 334% for microcel-

lular). Finally, the specific toughness, strain-at-break, and

specific strength of microcellular samples were found to be

lower than their solid counterparts.

Introduction

There is growing interest in developing biobased and

biodegradable polymers to help reduce dependency on

petroleum-based polymers, reduce the accumulation of

persistent plastic waste, and better control the emission of

CO2 in the environment [1–5]. Polylactide (PLA) is pop-

ular among biopolymers due to its close proximity in

properties to some of the synthetic polymers and its com-

mercial availability at a relatively low cost [6]. In addition

to its biobased and biodegradable attributes, PLA is bio-

compatible; thus, it can be used for biomedical applications

such as bone plates, bone screws, tissue repair, and drug

delivery [7–15]. In non-biomedical applications, PLA is

mainly used in packaging [16]. Despite its many advanta-

ges, PLA has a relatively low toughness and a narrow

processing window, which limit its widespread applica-

tions in areas such as structural components and high-end

packaging. For this reason, researchers around the world

have been investigating various techniques to improve the

toughness and processability of PLA.

The brittleness of PLA can be reduced by copolymeri-

zation [17–26], blending with other tough polymers [27–

S. Pilla � S. Gong

Department of Mechanical Engineering, University

of Wisconsin-Milwaukee, Milwaukee, WI, USA

S. Gong

Department of Materials, University

of Wisconsin-Milwaukee, Milwaukee, WI, USA

S. Gong (&)

Department of Biomedical Engineering, University

of Wisconsin, Madison, WI, USA

e-mail: shaoqingong@wisc.edu

A. Kramschuster � J. Lee � L.-S. Turng (&)

Department of Mechanical Engineering, Polymer Engineering

Center, University of Wisconsin, Madison, WI, USA

e-mail: turng@engr.wisc.edu

C. Clemons

Forest Products Laboratory, USDA Forest Service, Madison,

WI, USA

123

J Mater Sci (2010) 45:2732–2746

DOI 10.1007/s10853-010-4261-6

35], plasticization [36, 37], and by using special additives

such as hyperbranched polymers [38–42]. Although copo-

lymerization can be effective in improving the PLA

toughness, commercializing a new copolymer is often a

long and costly process [43], making them a less viable

option when considering large-scale applications of PLA.

The major drawback of plasticizers is that during long-term

use of the material, they have a tendency to migrate to the

surface of the material, causing brittleness [39]. Blending

PLA with a tough polymer may improve its toughness

effectively, but this generally requires a sufficiently high

amount of the tough polymer that can lead to a significant

reduction in modulus and strength. Recent studies indicate

that hyperbranched polyesters (HBPs) may provide a

plausible option to significantly increase the toughness of

PLA at a relatively low loading level [38–42]. This is due

to the very nature of HBPs, which fall under the category

of dendritic polymers and have unique characteristics

encompassing highly branched structures with a large

number of peripheral functionality [44]. HBPs can be used

for a variety of applications such as compatibilizers and

toughening agents for plastics, and drug nanocarriers for

biomedical applications [45–48]. In this study, HBPs are

used as tougheners to improve the toughness of PLA while

helping to control the cell morphology in the microcellular

PLA [38, 39]. Thus far, all the studies on the modification

of PLA with HBP have reported the use of conventional

injection molding to process the materials [38, 39]. How-

ever, in this study, we used the unique microcellular

injection-molding process to produce microcellular com-

ponents based on the PLA–HBP material systems.

The microcellular technology was first reported during

1970s and 1980s by Skripov and co-workers [49] and Suh

and co-workers [50]. The microcellular injection molding

process takes place in three steps: nucleation, cell growth,

and cell stabilization. First, the supercritical fluid (SCF),

such as nitrogen or carbon dioxide, acting as the blowing

agent, is dissolved into a polymer melt to form a single-

phase polymer–gas solution, that is, the polymer melt is

super-saturated with the blowing agent. Then, the pressure

is suddenly lowered to a value below the saturation pres-

sure triggering a thermodynamic instability and inducing

cell nucleation. Cell growth is controlled by the gas dif-

fusion rate and the stiffness of the polymer–gas solution. In

general, cell growth is affected by the following factors

[51]: (a) time allowed for cells to grow; (b) state of

supersaturation; (c) hydrostatic pressure applied to the

polymer; (d) temperature of the system; and (e) viscoelastic

properties of the single-phase polymer–gas solution. Other

than processing parameters, materials formulations such as

fillers and polymer blends also have strong influence on

cell nucleation and growth. Especially, addition of fillers,

which act as nucleating agents, leads to heterogeneous cell

nucleation. They provide a large number of nucleation sites

leading to higher cell densities and smaller cell sizes [51–

59]. A detailed review of the microcellular process is

presented in [60].

The employment of SCF such as nitrogen as used in this

study reduces the viscosity of the polymer melt [61–63]

due to the formation of a single-phase polymer–gas solu-

tion, enabling the polymer to be processed at lower tem-

peratures and pressures [64, 65]. This is a very desirable

feature for biobased polymers such as PLA, which are

moisture and heat sensitive. Microcellular plastics are

characterized by cell densities on the order of 107–

109 cells/cm3 and cell sizes on the order of several tens of

microns or less. Compared to microcellular plastics, con-

ventional foamed plastics possess relatively lower cell

densities (in the range of 103–106 cells/cm3) and larger cell

size (on the order of 100 lm or more), thus leading to

inferior material properties [66]. Compared with the con-

ventional injection molding process, the microcellular

injection-molding process produces components with

increased dimensional stability, less thermal degradation,

and less material [52–54].

This study investigated PLA–HBP blends produced via

microcellular injection molding using two types of HBP

polymers. Poly (maleic anhydride-alt-1-octadecene) was

used as a cross-linking agent for the HBP. Specimens have

also been produced using conventional injection molding

to compare with foamed specimens. Finally 2% nanoclay

was used to study the effect of nanofillers on the properties

of PLA–HBP blends. More information on PLA–nanoclay

nanocomposites can be found in [52, 67, 68].

Experimental

Materials

Polylactide (NatureWorksTM PLA 3001D) in pellet form

was obtained from NatureWorks� LLC (Minnetonka, MN).

It has a specific gravity of 1.24 and a melt flow index around

15 g/10 min. PLA 3001D was synthesized from approxi-

mately 92% L-lactide and 8% meso-lactide [69]. The HBPs,

under the trade names Boltorn H20� and Boltorn H2004�

were provided by Perstorp Speciality Chemicals AB,

Sweden. Boltorn H20� has a nominal molecular weight of

1750 g/mol and consists of 16 primary hydroxyl (OH–)

groups; Boltorn H2004� has a nominal molecular weight of

3100 g/mol and possesses six hydroxyl (OH–) groups. Poly

(maleic anhydride-alt-1-octadecene) (PA) obtained from

Sigma–Aldrich�, was used as a cross-linking agent for

the HBPs. It has a number average molecular weight of

30,000–50,000 g/mol and is a copolymer of octadecene and

maleic anhydride. The organically modified montmorillonite

J Mater Sci (2010) 45:2732–2746 2733

123

(MMT) nanoclay, Cloisite�30B, was supplied by Southern

Clay Products, Inc. The nanoclay was surface treated by an

ion exchange reaction between Na? existing in the gallery

of the nanoclay and quaternary ammonium cations. Cloi-

site�30B was modified with bis-(2-hydroxyethyl) methyl

(hydrogenated tallowalkyl) ammonium cations.

Methods

Processing

The PLA was combined with PA, HBP (Boltorn H20� and

Boltorn H2004�), and nanoclay (Cloisite�30B) (used as

received) into a variety of formulations, with Naugard-10

and Naugard-524 introduced as antioxidants. Table 1 pre-

sents the formulations compounded, injection–molded, and

evaluated in this study. PA was used in the formulations

because previous study has indicated that greater tough-

ening effect can be achieved due to the formation of a

network between HBP and PA resulting from a reaction

between the HBPs’ hydroxyl groups and the PA’s anhy-

dride groups during the melt compounding process [38].

Also among the HBPs, Boltorn H20� has 16 hydroxyl

(–OH) functional groups and Boltorn H2004� has six

hydroxyl (–OH) functional groups. Therefore, based on

their reactivity with polyanhydride (PA), required amounts

of HBPs have been computed. Therefore, though the

overall percentage of the HBP ? PA remains the same, the

constituent weights vary due to the different amount of

reactive groups on the periphery of individual HBPs.

Prior to compounding and injection-molding, several

master batches were created using a thermokinetic mixer

(k-mixer). All the master batches were compounded in

quantities of 200 g. The rotor speed was 4000 rpm, and the

discharge temperature was either 175 �C (for master bat-

ches containing PLA with PA, the antioxidants, and the

nanoclay, if used) or 80 �C (for a master batch consisting

of Boltorn H20� and PA for Experiment #5). After dis-

charging, the molten blend was pressed into a flat sheet,

and subsequently granulated. The contents of the master

batches created are provided in Table 2. For Experiments 1

and 5 (Table 2), master batches of PLA, Naugard-10, and

Naugard-524 were created. For Experiments #2 and 3,

master batches of PLA, PA, Naugard-10, and Naugard-524

were created, while for Experiment #4, Cloisite�30B

nanoclays were added. Boltorn H2004� comes in a liquid

form and was added to these formulations during twin-

screw compounding using a Cole-Parmer peristaltic pump

(model #7553–80).

All of these master batches were then let-down with PLA

at the appropriate ratio to achieve the formulations descri-

bed in Table 1. The co-rotating, twin-screw compounding

extruder had a screw diameter of 32 mm and an L/D ratio of

36.25. The materials were all fed into the main feed throat

and extruded at 11.4 kg/h (the pumping rate of the peri-

staltic pump was set at 8.5 g/min for Experiment #2, and

17 g/min for Experiments #3 and #4) with a screw speed of

200 rpms and a melt temperature of approximately 170–

180 �C to create the formulations described in Table 1.

Tensile bars (ASTM D638-03, Type I) were injection-

molded using an Arburg Allrounder 320S (Lossburg,

Germany) with a 25-mm diameter screw and equipped with

microcellular (under the trade name MuCell�) technol-

ogy (Trexel, Inc., Woburn, MA). Solid and microcellular

tensile bars were molded at the processing conditions

Table 1 Percent composition of the materials compounded

Experiment Sample PLA PA HBP Naugard-10

(0.2 wt%

total formulation)

Naugard-524

(0.2 wt%

total formulation)

Cloisite�30B

1 PLA 99.6 0.0 0 0.2 0.2 0

2 PLA–6%(H2004 ? PA) 93.6 1.5 4.5 0.2 0.2 0

3 PLA–12%(H2004 ? PA) 87.6 3.0 9.0 0.2 0.2 0

4 PLA–12%(H2004 ? PA)–2%NC 85.6 3.0 9.0 0.2 0.2 2

5 PLA–12%(H20 ? PA) 87.6 7.4 4.6 0.2 0.2 0

Table 2 Master batches created for compounding

Master batch PLA PA HBP Naugard-10 Naugard-524 Cloisite�30B

Exps 1, 5 X X X

Exps 2, 3 X X Added during twin screw extrusion X X

Exp 4 X X Added during twin screw extrusion X X X

2734 J Mater Sci (2010) 45:2732–2746

123

indicated in Table 3. The processing temperatures for both

the extrusion and injection molding processes were kept

between 170 and 180 �C to avoid significant PLA thermal

degradation. Note that with conventional injection-mold-

ing, supercritical nitrogen fluid (SCF) was not introduced

into the material. In addition, the pack/hold stage is absent

in microcellular injection molding due to the homogeneous

packing pressure that results from the nucleation and

growth of cells [65], which was shown to drastically reduce

the shrinkage and warpage of microcellular injection-

molded parts when compared with conventional injection

molding, thereby greatly enhancing the dimensional sta-

bility of molded parts with a complex geometry [55]. A

slight variation in the wt% SCF content is observed for the

microcellular injection-molded samples. This can be

attributed to varying shot weights between the neat resin

and the composites. This shot weight is used to compute

the wt% SCF, shown in Eq. 1 [53]

wt%SCF ¼ _mtð27:8Þm

ð1Þ

where _m is the mass flow rate of the SCF (kg/h), t is the

SCF dosage time (s), m is the shot weight (g), and 27.8 is a

conversion factor.

Wide-angle X-ray diffraction

WAXRD analysis was performed on Scintag XDS 2000

with Ni-filtered Cu Ka radiation (1.5418 A�) at room

temperature in the range of 2h = 1.5–9� with a scanning

rate of 1�/min.

Differential scanning calorimetry (DSC)

A differential scanning calorimeter (TA Instruments, DSC-

Q20) was used to study the thermal properties of the blend

materials. From each sample, 8–10 mg was sliced from the

injection-molded specimens and placed in a hermetically

sealed aluminum pan under 50 mL/min nitrogen flow. The

samples were first heated from 40 �C to 180 �C and then

subjected to an isothermal stage for 3 min, cooled to 0 �C,

and finally reheated to 200 �C. The ramp speed in all of the

heating and cooling processes was 10 �C/min.

The degree of crystallinity of PLA was computed using

Eq. 2 [70]:

vc % Crystallinityð Þ ¼ DHm

DH0m

� 100

wð2Þ

where DHm is the enthalpy for melting, DH0m is the

enthalpy of melting for a 100% crystalline PLA sample,

which is 93.7 J/g, and w is the weight fraction of PLA in

the sample.

In order to determine the degree of crystallinity of the

sample, the extra heat absorbed by the crystallites formed

during heating (i.e., cold crystallization) had to be sub-

tracted from the total endothermic heat flow due to the

melting of the total amount of crystallites [71]. Thus, the

modified equation can be written as follows:

vc % Crystallinityð Þ ¼ DHm � DHcc

DH0m

� 100

wð3Þ

where DHcc is the cold crystallization enthalpy.

Scanning electron microscopy (SEM)

The fracture surfaces obtained from the tensile tests were

examined using SEM (Hitachi S-570) operated at 10 kV.

All the specimens were sputter coated with a thin layer of

gold (*20 nm) prior to examination.

Transmission electron microscopy (TEM)

The structure of the PLA–HBP blends were investigated

using a Hitachi H-600 TEM operated at 75 kV. The ultra-

thin sections with a thickness of *70 nm were micro-

tomed at room temperature (*25 �C) using an RMC

ultramicrotome (Model# MT-7000). No stain was used.

The TEM sections were taken from the middle portion of

the tensile test specimens.

Dynamic mechanical analysis (DMA)

The dynamic mechanical analysis measurements were

performed on a TA Q800 DMA instrument in single

cantilever mode. The dimensions of the rectangular

samples were about 17.6 9 12.7 9 3.2 mm3. Samples

were heated at a rate of 3 �C/min ranging from 0 �C to

85 �C with a frequency of 1 Hz and a strain of 0.02%,

which is in the linear viscoelastic region, as determined

by a strain sweep.

Table 3 Injection-molding conditions used to mold the tensile bars

(S solid, M microcellular)

S M

Mold temperature (�C) 20 20

Nozzle temperature (�C) 175 170

Injection speed (cm3/s) 20 20

wt% SCF content n/a 0.56

Pack pressure (bar) 795 –

Pack time (s) 7.5 –

Screw recovery speed (RPM) 280 280

Cooling time (s) 35 35

Microcellular process pressure (bar) n/a 190

J Mater Sci (2010) 45:2732–2746 2735

123

Tensile testing

The static tensile properties (i.e., modulus, strength, strain-

at-break, and toughness) were measured at room temper-

ature (*25 �C) and atmospheric conditions (relative

humidity of *50 ± 5%) with a 50-kN load cell on an

Instron Model 3369 tensile tester. The crosshead speed was

set at 5 mm/min. All the tests were carried out according to

the ASTM standard (ASTM-D638); five specimens of each

sample were tested and the average results were reported.

Results and discussions

WAXRD analysis

The structure of the solid and microcellular PLA–

12%(H2004 ? PA)–2%Nanoclay composite was studied

using wide angle X-ray diffraction (WAXRD). The

WAXRD pattern provides a convenient way to determine

the degree of clay dispersion based on the diffraction angle

(2h), shape, and intensity of the (001) diffraction peak from

the nanoclays dispersed in the polymer matrix. In general,

when the nanoclay are fully exfoliated the X-ray diffraction

peaks disappear, but if the nanoclay is intercalated then the

diffraction peak shifts to lower angles due to the increase in

the interlayer gallery spacing (d001).

Figure 1 shows the WAXRD pattern of nanoclay and

PLA–12%(H2004 ? PA)–2%Nanoclay composites (both

solid and microcellular). For reference, the WAXRD pat-

terns of solid and microcellular PLA–12%(H2004 ? PA)

without nanoclay are also presented. As shown in the figure,

the diffraction peaks from the nanoclays in both the solid

and microcellular PLA–12%(H2004 ? PA)–2%Nanoclay

composites shifted to lower angles. The diffraction peak

angle (2h) for the (001) peak of the as-received nanoclay

(Cloisite�30B) was 4.74�. For this diffraction angle, the

basal spacing for the nanoclay, as per the Bragg’s diffrac-

tion equation, is d001 = 1.86 nm. This compares reasonably

well with the information provided by the manufacturer.

For both the solid and microcellular PLA–12%(H2004 ?

PA)–2%Nanoclay composites, the diffraction peaks were

observed at 2h = 3.9�, and the corresponding basal spacing

was calculated to be d001 = 2.3 nm. As stated above, an

increase in the interlayer basal spacing is indicative of an

intercalated structure. Thus, as noted in the previous stud-

ies, the PLA molecules intercalated into the clay layers

during the melt compounding process and increased the

interlayer spacing [54], but the clay was not exfoliated.

A visual examination of this inference was made using

TEM, as discussed in the next section.

Morphological properties

Microstructure analysis via TEM

Figure 2a, b shows the TEM images of the solid and

microcellular PLA–12%(H2004 ? PA)–2%Nanoclay com-

posites, respectively. The dark lines are the clay particles

and bright area is the PLA matrix. The degree of nanoclay

dispersion was similar in the solid and microcellular

specimens. The degree of nanoclay dispersion generally

depends on the interaction between the nanoclays and the

polymer matrix. The TEM images clearly showed that

nanoclays were dispersed quite uniformly in the polymer

matrix, mostly showing an intercalated structure occa-

sionally with a few larger nanoclay agglomerates observed

here and there. The inset in Fig. 2c shows an enlarged view

of the microcellular PLA–12%(H2004 ? PA)–2%Nano-

clay composite with an intercalated structure. These results

agree with the XRD results shown in Fig. 1.

Figure 3 shows the TEM images of the solid and

microcellular PLA–HBP blends. The dark particles (shown

by black arrows) and the dark lines are (HBP ? PA) and

the bright area is the PLA matrix. The tiny microcells in the

microcellular samples are indicated by white arrows. Fig-

ure 3a, b shows the TEM images of the solid and micro-

cellular PLA–6%(H2004 ? PA), respectively, in which

the average size of the individual (H2004 ? PA) parti-

cles ranged from approximately 10–100 nm. Figure 3c, d

shows the TEM images of solid and microcellular PLA–

12%(H2004 ? PA), respectively. Compared with PLA–

6%(H2004 ? PA), a much more homogeneous (H2004 ?

PA) distribution was achieved in the PLA–12%(H2004 ?

PA) sample. Figure 3e, f shows, respectively, the TEM

images of solid and microcellular PLA–12%(H2004 ?

PA)–2%Nanoclay composites. In this case, the dark

regions indicate both the (HBP ? PA) and nanoclays.

Fig. 1 WAXRD patterns of Nanoclay (30B), solid and microcellular

PLA–12%(H2004 ? PA) and PLA–12%(H2004 ? PA)–2%Nano-

clay composites

2736 J Mater Sci (2010) 45:2732–2746

123

Finally, the TEM images of PLA–12%(H20 ? PA) are

shown in Fig. 3g, h. Similar to the PLA–12%(H2004 ?

PA) sample, both dark particles and dark lines are shown in

the TEM images of PLA–12%(H20 ? PA); however, the

overall dispersion of the 12%(H20 ? PA) in the PLA was

less homogenous than that of the 12%(H2004 ? PA) in the

PLA.

Fracture surface analysis via SEM

Figure 4 shows representative SEM images of the solid

(images on the left) and microcellular (images on the right)

PLA and PLA–HBP blends. All images were taken at the

same magnification of 9100 (scale bar: 100 lm). The SEM

images provide information on the microstructure, includ-

ing the cell morphology in microcellular specimens and the

fracture behaviors of the specimens [52–57, 64, 65, 72].

The fracture surface of the solid PLA (Fig. 4a) was

smooth without obvious plastic deformation indicating

brittle behavior. For the solid PLA–HBP blends, the frac-

ture surfaces (Fig. 4c, e, g, i) were much rougher, indi-

cating plastic deformation. In general, the microcellular

specimens showed similar morphological deformation as

that of the solid samples.

The average cell size and cell density were quantita-

tively determined using an image analysis tool (UTHSCSA

ImageTool).

The cell density was calculated using [73]:

Cell density ¼ N

L2

� �3=2

M ð4Þ

where N is the number of cells, L is the linear length of the

area, and M is a unit conversion factor resulting in the

number of cells per cm3.

The graphical results of the average cell size and cell

density of the microcellular samples are shown in Fig. 5. The

left set of plots show the effect of (H2004 ? PA) on cell size

and density with variation in the (H2004 ? PA) content and

composition (with or without nanoclay) and the right set of

plots shows the effects of addition of different types of HBPs

(H20 vs. H2004) on cell size and density. The average

cell size of pure PLA was 39 lm. With the addition of

6%(H2004 ? PA), the average cell size decreased to 31 lm.

Increasing the content of (H2004 ? PA) to 12%, that is for

the PLA–12%(H2004 ? PA) sample, the average cell size

decreased to 16 lm and with the addition of 2% nanoclay,

that is, for the PLA–12%(H2004 ? PA)–2%Nanoclay com-

posite, the average cell size reduced to 10 lm. However,

with the addition of 12%(H20 ? PA), that is for the PLA–

12%(H20 ? PA), the average cell size decreased to 31 lm,

which was higher than 16 lm achieved for PLA–

12%(H2004 ? PA).

The cell density increased by about 1.2 and 5 times for the

PLA–6%(H2004 ? PA) and PLA–12%(H2004 ? PA)

samples compared with pure PLA. Furthermore, with the

addition of 2% nanoclay, that is for the PLA–12%

(H2004 ? PA)–2%Nanoclay composite, the cell density

increased 10 times compared with that of the microcellular

pure PLA; however, for the PLA–12%(H20 ? PA), the cell

density increased by approximately 1.1 times. Thus, com-

pared with the microcellular PLA–12%(H2004 ? PA)

specimen, microcellular PLA–12%(H20 ? PA) showed

Fig. 2 TEM images of PLA–

12%(H2004 ? PA)–

2%Nanoclay composite:

a PLA–12%(H2004 ? PA)–

2%Nanoclay (Solid),

b PLA–12%(H2004 ? PA)–

2%Nanoclay (Microcellular),

c enlarged view of the PLA–

12%(H2004 ? PA)–

2%Nanoclay (Microcellular)

J Mater Sci (2010) 45:2732–2746 2737

123

Fig. 3 TEM images of solid and microcellular PLA–HBP blends:

a PLA–6%(H2004 ? PA) (Solid), b PLA–6%(H2004 ? PA)

(Microcellular), c PLA–12%(H2004 ? PA) (Solid), d PLA–12%

(H2004 ? PA) (Microcellular), e PLA–12%(H2004 ? PA)–2%

Nanoclay (Solid), f PLA–12%(H2004 ? PA)–2%Nanoclay (Micro-

cellular), g PLA–12%(H20 ? PA) (Solid), h PLA–12%(H20 ? PA)

(Microcellular)

Fig. 4 Representative SEM images of the fracture surfaces of solid and

microcellular PLA and PLA–HBP blends: a pure PLA (Solid), b pure

PLA (Microcellular), c PLA–6%(H2004 ? PA) (Solid), d PLA–6%

(H2004 ? PA) (Microcellular), e PLA–12%(H2004 ? PA) (Solid),

f PLA–12%(H2004 ? PA) (Microcellular), g PLA–12%(H2004 ?

PA)–2%Nanoclay (Solid), h PLA–12%(H2004 ? PA)–2%Nanoclay

(Microcellular), i PLA–12%(H20 ? PA) (Solid), j PLA–12%(H20 ?

PA) (Microcellular)

2738 J Mater Sci (2010) 45:2732–2746

123

inferior cell structure (larger cell size and lower cell density).

Therefore, it is evident that the effect of H20 ? PA on the

cell structure of PLA was meager. On the other hand, it can

be inferred that the H2004 ? PA had a positive effect on the

cellular morphology of PLA. Moreover, addition of nano-

clay increased the cell density almost 10 times and decreased

the cell size by 75%. This increase in cell density and

decrease in cell size can be attributed to: (1) more cells being

nucleated in the presence of nanoclay resulting in less SCF

available for each cell to grow, thereby reducing cell size; or

(2) increase in melt viscosity due to the addition of nanoclay

thereby inducing strain hardening and thus hindering cell

growth and coalescence, and reducing cell size [74]. These

results agree with the findings from the literature that addi-

tion of fillers decreases cell size and increases cell density

[58, 59, 65, 75]. Thus, nanoclay acted as a nucleating agent,

thereby promoting heterogeneous cell nucleation and led to

more uniform cell nucleation and growth behavior.

Finally, it should be noted that cell size and distribution

are not always uniform due to the dynamic nature of the

microcellular injection-molding process. The quantitative

analyses presented in Fig. 5 include representative values

taken from the center portion of the cross section of the

tensile bars. However, cell size and cell density varies

throughout the thickness of the part due to shear and rapid

cooling at the polymer–mold interface, as well as other

phenomena. More specifically, there is a typical solid skin

layer near the polymer–mold interface where cells are not

visible due to rapid cooling of the material, which ham-

pered cell growth.

Thermal properties

The thermal properties of PLA and PLA–HBP blends,

including crystallization and melting behavior, were inves-

tigated using DSC. The thermograms and the numerical

values of temperature and enthalpy obtained from the first

and second heating cycles are shown in Fig. 6 and Table 4,

respectively. The data obtained from the first heating cycle

provide the thermal history of the injection-molded samples,

while the data obtained from the second heating cycle allows

for a direct comparison of the crystallization behavior of

different materials after erasing the thermal history through

the first heating cycle.

First heating cycle

As shown in Fig. 6a, an endothermic peak exists near

the glass transition phase of solid PLA, solid PLA–

12%(H20 ? PA), microcellular PLA–6%(H2004 ? PA),

microcellular PLA–12%(H2004 ? PA) and microcellular

PLA–12%(H2004 ? PA)–2%Nanoclay composite. This is

due to physical aging of the polymeric materials [76–78]

and is related to the inherent distribution of the relaxation

times of polymer chains [79]. Also, two crystallization

peaks were observed for all the solid and microcellular

Fig. 5 The average cell size and cell density of microcellular PLA

and PLA–HBP blends

Fig. 6 Melting curves of solid and microcellular PLA and PLA–

HBP blends. Data obtained from a first heating run and b second

heating run; (a) pure PLA, (b) PLA–6%(H2004 ? PA), (c) PLA–

12%(H2004 ? PA), (d) PLA–12%(H2004 ? PA)–2%Nanoclay, (e)

PLA–12%(H20 ? PA)

J Mater Sci (2010) 45:2732–2746 2739

123

specimens. The first peak is referred to as cold-crystalli-

zation peak and the second peak as recrystallization peak.

With the addition of (H2004 ? PA) (6% or 12%), the peak

temperatures of the cold-crystallization peaks decreased.

This indicates that the addition of (H2004 ? PA) promoted

initial cold crystallization of the PLA material. Moreover,

with the addition of 2% nanoclay, a similar decreasing

trend in the cold-crystallization peak temperature was

observed. On the contrary, the addition of 12%(H20 ? PA)

increased the cold-crystallization peak temperature, indi-

cating that the crystallization process was partially delayed

during the heating cycle. The recrystallization peaks, which

occurred due to the restructuring of certain existing crys-

talline structures at higher temperatures [5], appeared just

before the melting peaks for all the solid and microcellular

specimens.

Table 4 shows the numerical values of temperatures and

enthalpies from the first heating curve of the PLA and

PLA–HBP blends. The enthalpies of the cold-crystalliza-

tion peaks decreased with the addition of 6% and

12%(H2004 ? PA) HBPs, indicating that there was

enhanced PLA crystallization during the cooling during

injection molding. As a result, there was a higher crystal-

linity and a less cold crystallization. Moreover, with the

addition of 2% nanoclay, the enthalpy of cold crystalliza-

tion also decreased, indicating further enhancement of

crystallization during cooling. Indeed, as shown in Table 4,

the degree of crystallinity (computed using Eq. 3) of all the

PLA–HBP blends was observed to be higher than that of

the pure PLA. This agrees with the findings reported in the

literature that addition of HBP enhanced the degree of

crystallinity in PLA [39, 42]. Furthermore, nanoclay acted

Table 4 Thermal characteristics of PLA and PLA–HBP blends

Sample # Cold crystallization Recrystallization Melting Degree of

crystallinity

Temp (�C) Enthalpy (J/g) Temp (�C) Enthalpy (J/g) Temp (�C) Enthalpy J/g v (%)

Solid (first heating)

PLA 105.2 -27.6 155.5 -0.7 169.1 34.9 8

PLA–6%(H2004 ? PA) 92.1 -24.8 153.6 -1.9 169.2 36.7 11

PLA–12%(H2004 ? PA) 91.4 -22.6 154.2 -2.1 170.8 36.4 14

PLA–12%(H2004 ? PA)–2%NC 90.6 -20.3 153.5 -1.4 170.5 35.8 18

PLA–12%(H20 ? PA) 107.2 -23.3 154.9 -0.7 171.9 34.6 13

PLA–12%(H2004 ? PA) 91.4 -22.6 154.2 -2.1 170.8 36.4 14

Microcellular (first heating)

PLA 101.9 -24.6 155.4 -1.7 169.4 35.1 9

PLA–6%(H2004 ? PA) 101.1 -24.7 155.0 -1.1 170.2 35.7 11

PLA–12%(H2004 ? PA) 98.3 -24.4 154.8 -1.7 171.1 37.1 13

PLA–12%(H2004 ? PA)–2%NC 94.7 -19.8 154.2 -1.5 170.7 36.1 18

PLA–12%(H20 ? PA) 105.2 -22.2 154.7 -0.4 169.9 33.7 13

PLA–12%(H2004 ? PA) 98.3 -24.4 154.8 -1.7 171.1 37.1 13

Solid (second heating)

PLA 109.1 -26.8 – – 162.8 169.3 36.4 10

PLA–6%(H2004 ? PA) 104.5 -28.7 – – 163.9 169.6 39.1 12

PLA–12%(H2004 ? PA) 104.3 -27.1 – – 164.6 169.8 38.4 14

PLA–12%(H2004 ? PA)–2%NC 103.9 -25.2 – – 165.1 170.4 38.7 17

PLA–12%(H20 ? PA) 111.2 -27.2 – – 165.4 169.2 37.6 13

PLA–12%(H2004 ? PA) 104.3 -27.1 – – 164.6 169.8 38.4 14

Microcellular (second heating)

PLA 108.1 -26.6 – – 163.2 169.1 36.9 11

PLA–6%(H2004 ? PA) 104.3 -28.2 – – 164.1 169.5 38.5 12

PLA–12%(H2004 ? PA) 103.6 -26.8 – – 164.2 169.6 38.8 15

PLA–12%(H2004 ? PA)–2%NC 102.4 -22.2 – – 164.3 170.4 35.6 17

PLA–12%(H20 ? PA) 110.7 -27.1 – – 165.1 169.1 37.4 13

PLA–12%(H2004 ? PA) 103.6 -26.8 – – 164.2 169.6 38.8 15

2740 J Mater Sci (2010) 45:2732–2746

123

as crystallization nucleating agents and further increased

degree of crystallinity [52, 53, 56, 80, 81]. Although the

cold crystallization peak temperature was higher in the

PLA–12%(H20 ? PA) specimen, the degree of crystallin-

ity in this sample was similar to the PLA–12%(H2004 ?

PA) specimen. Among all the solid and microcellular

samples, the degree of crystallinity was observed to be the

highest (18% for both solid and microcellular) for PLA–

12%(H2004 ? PA–2%Nanoclay composite. Finally, no

significant difference was observed between the degree of

crystallinity of the corresponding solid and microcellular

samples.

Second heating cycle

Figure 6b and Table 4 show the thermogram and numeri-

cally analyzed data of PLA and PLA–HBP blends,

respectively, from the second heating cycle. Unlike in the

first heating cycle, no endotherm peaks were observed near

the glass transition temperature (Tg) because the enthalpic

recovery that occurred during the first heating cycle is

kinetic in nature [5]. Also, only one crystallization peak,

that is, cold-crystallization was observed. Similar to the

first heating cycle, the addition of (H2004 ? PA) (6% or

12%) decreased the peak temperatures of the cold-crys-

tallization peaks. This indicates that the addition of H2004

promoted the cold crystallization of the PLA material.

Again, as previously observed, for the PLA–12%(H2004 ?

PA)–2%Nanoclay composite, the cold-crystallization peak

temperature did not change. Also, as in the case of the first

heating cycle, the addition of 12%(H20 ? PA) increased

the cold-crystallization peak temperature of PLA by a few

degrees. The degree of crystallinity of all the samples was

found to be either the same or slightly higher than that

obtained during the first heating cycle.

Double melting peaks were obtained for all the speci-

mens during the second heating cycles (Fig. 6b; Table 4).

This is commonly observed [82] and, in our case, may be

due to the differences in crystalline morphology (e.g.,

lamellar thickness or spherulite size) from melt and cold

crystallization, for instance, which may have slightly dif-

ferent melt temperatures [83].

Weight reduction of the microcellular samples

The weight of PLA without HBP and PA was reduced by

16% using microcellular injection molding. Addition of

12%(H20 ? PA) yielded blends with the least weight

reduction (10%). Though the weight was only reduced by

11% in blends with 6%(H2004 ? PA), doubling the

additive content yielded similar weight reductions (15%) as

with the pure PLA (16%). When 2% nanoclay was added to

the PLA–12%(H2004 ? PA) blend, the weight reduction

dropped to 13%.

Overall, the variation in weight reduction among all the

samples was due to the effects of the blend composition on

the behavior of cell nucleation and growth, different

pressure-specific volume–temperature (pvT) properties that

affect the melt properties of these materials and the pro-

cessing parameters such as SCF dosage time, and the

dynamic nature of the microcellular injection molding.

Dynamic mechanical properties (DMA)

The viscoelastic properties of PLA and the PLA–HBP

blends were studied using DMA. The DMA properties

reported here are the actual properties measured for the

solid and microcellular specimens without taking into

account the weight reduction of the microcellular speci-

mens. In general, a declining trend was observed for the

storage moduli of all the solid and microcellular specimens

with respect to temperature with the most rapid reduction

occurring at the glass transition region (Fig. 7). In the glassy

region (\55 �C), the storage modulus of the solid specimen

Fig. 7 Storage moduli of solid and microcellular PLA and PLA–HBP

blends: (a) pure PLA, (b) PLA–6%(H2004 ? PA), (c) PLA–

12%(H2004 ? PA), (d) PLA–12%(H2004 ? PA)–2%Nanoclay,

(e) PLA–12%(H20 ? PA)

J Mater Sci (2010) 45:2732–2746 2741

123

decreased with increasing (H2004 ? PA). Addition of 2%

nanoclay increased the storage modulus of the solid speci-

men (i.e., PLA–12%(H2004 ? PA)–2%Nanoclay com-

posite), but it was still lower than that of the solid PLA.

Similarly, addition of 12%(H20 ? PA) decreased the

storage modulus of the solid specimen. Furthermore, addi-

tion of H2004 reduced storage modulus more than that due

to the addition of H20 at the same 12% loading level. This is

probably due to the larger number of –OH groups (16) on

the periphery of H20 compared with H2004 (6), resulting in

a higher cross-link density after reacting with PA. In the

glass transition region, no cross-over was observed in any

solid samples, and the same trend of storage modulus as

seen in the glassy region continued. Above the glass tran-

sition region (temperatures above 70 �C), no significant

difference was observed in the storage moduli of all the

solid specimens.

For the microcellular specimens (Fig. 7), in general, the

storage modulus followed a similar trend as that of the

solid samples, except that in the glassy region, the differ-

ence between the storage modulus of microcellular PLA

and microcellular PLA–6%(H2004 ? PA) was very mini-

mal. Also, the storage modulus of microcellular PLA–

12%(H2004 ? PA)–2%Nanoclay composite was found to

be lower than that of microcellular PLA–12%(H2004 ?

PA). In the glass transition region, a cross-over was

observed between the microcellular PLA and microcellu-

lar PLA–6%(H2004 ? PA). The slightly different trend

observed in the solid and microcellular samples may be

attributed to the variation of weight reduction and cell

morphology in these microcellular specimens. Above the

glass transition region, no significant difference was

observed in the storage moduli of all the microcellular

specimens. Finally, the storage moduli of the microcellular

samples was found to be higher than their solid counter-

parts due to the introduction of microcells [52].

The glass transition temperatures (Tg) obtained from the

peaks of the Tan-d curves are tabulated in Table 5. The Tg

of all solid PLA–HBP blends were lower than that of the

solid PLA. This is consistent with previously published

results [38, 40]. However, the Tg of the microcellular PLA–

HBP blends was higher than that of the microcellular PLA.

Also, the Tg of the microcellular specimens were found to

be lower than their solid counterparts, which agrees with

our previous results [52, 53, 57]. This might be due to

the increased molecular mobility due to the presence of

microcells in the microcellular specimens.

Figure 8 shows the area under the tan-d curves of all the

solid and microcellular samples. In general, a larger area

underneath the tan-d peak indicates that the molecular chains

exhibit a higher degree of mobility and, therefore, have

increased damping capability [84]. As shown in Table 5, the

Table 5 The glass transition

temperatures and area

integration under the tan-dcurves of solid and

microcellular PLA and PLA–

HBP blends measured using

DMA

Samples Tg (�C) Area under the tan-d curve

(cm2)

Solid Microcellular Solid Microcellular

PLA 73.2 67.5 73.2 67.5

PLA–6%(H2004 ? PA) 70.7 69.9 70.7 69.9

PLA–12%(H2004 ? PA) 70.3 69.3 70.3 69.3

PLA–12%(H2004 ? PA)–2%NC 70.9 69.1 70.9 69.1

PLA–12%(H20 ? PA) 72.1 71.8 72.1 71.8

Fig. 8 Tan-d curves of solid and microcellular PLA and PLA–HBP

blends: (a) pure PLA, (b) PLA–6%(H2004 ? PA), (c) PLA–12%

(H2004 ? PA), (d) PLA–12%(H2004 ? PA)–2%Nanoclay, (e) PLA–

12%(H20 ? PA)

2742 J Mater Sci (2010) 45:2732–2746

123

area under the tan-d curve increased with the addition of 6%

or 12% (H2004 ? PA). Thus, the damping ability, that is, the

energy absorption capacity of PLA, improved with addition

of (H2004 ? PA). Also, the area under the tan-d curve

was similar between PLA–12%(H2004 ? PA) and PLA–

12%(H2004 ? PA)–2%Nanoclay. On the other hand, the

addition of (H20 ? PA) did not change the area under the

Tan-d curve of PLA. This shows that (H20 ? PA) did not

effectively toughen the PLA matrix.

Static mechanical properties

Figures 9 and 10 show the stress–strain plots and specific

mechanical properties of the PLA and the PLA–HBP

blends, respectively, obtained by tensile testing of the solid

and microcellular specimens. The specific properties such

as specific toughness, specific strength, and specific mod-

ulus were obtained by dividing the static properties with

the density of the respective material.

Considerable necking was observed for all the solid

and microcellular samples (Fig. 9). Solid PLA underwent

strain softening but addition of 6%(H2004 ? PA) induced

considerable cold drawing and ductile failure. A similar

trend of strain softening and cold drawing was observed

when the (H2004 ? PA) content was increased to 12%,

except that the cold drawing was greater than in the solid PLA–

6%(H2004 ? PA). Further, for the solid PLA–12%(H2004 ?

PA)–2%Nanoclay composite, the cold drawing was the

greatest among all of the samples indicating that the

nanoclay induced plastic deformation in solid PLA–

12%(H2004 ? PA). On the other hand, the addition of

(H20 ? PA) did not lead to higher ductility in the solid PLA.

The microcellular specimens showed a similar trend as

that of the solid samples, but the solid samples had a much

higher ductility than the microcellular specimens. This can

be attributed to the presence of certain large cells that

decreased the effective load bearing area and/or served as

stress concentrators in the microcellular samples [54].

Overall, among all solid and microcellular specimens,

PLA–12%(H2004 ? PA)–2%Nanoclay composite showed

the highest ductility.

The specific tensile properties of PLA and PLA–HBP

blends are shown in Fig. 10. The left set of plots show the

effect of (H2004 ? PA) content and nanoclay; the right set

of plots shows the effect of addition of different HBPs

(H20 and H2004) on various mechanical properties of the

material systems. Figure 10a shows the specific toughness

of PLA and PLA–HBP blends. Fracture toughness, which

is the energy-to-fracture per unit volume of the specimen,

is obtained by integrating the area under the stress–strain

curve [85]. Among the solid specimens, addition of

6%(H2004 ? PA) increased the toughness of solid PLA by

*240% (Fig. 10a). However, by increasing the content of

(H2004 ? PA) to 12%, the toughness increased to 275%.

This is a significant achievement in terms of toughness

increase. Further, owing to addition of 2% nanoclay to the

solid PLA–12%(H2004 ? PA)–2%Nanoclay composite,

the toughness increased to a gigantic value of 405%. This

finding agrees with those from previous studies which

showed that addition of two nanoscale fillers simulta-

neously can result in significantly enhanced properties [86].

Conversely, addition of 12% (H20 ? PA) decreased the

toughness by about 16% compared with solid PLA. This

might be due to the higher cross-linking density induced

between H20 and PA because it has more –OH groups (16)

on the periphery compared with H2004 (6). A similar trend

was observed for the microcellular specimens. Addition

of 6%(H2004 ? PA) resulted in *97% increase in the

toughness of microcellular PLA. Increasing the (H2004 ?

PA) content to 12%, the toughness increased to 191%.

Moreover, owing to the addition of 2% nanoclay, the

toughness of PLA–12%(H2004 ? PA)–2%Nanoclay com-

posite increased to about 334%. The decrease in the

toughness of microcellular PLA–12%(H20 ? PA) sample

Fig. 9 Tensile stress–strain curves of solid and microcellular PLA

and PLA–HBP blends: (a) pure PLA, (b) PLA–6%(H2004 ? PA),

(c) PLA–12%(H2004 ? PA), (d) PLA–12%(H2004 ? PA)–2%

Nanoclay, (e) PLA–12%(H20 ? PA)

J Mater Sci (2010) 45:2732–2746 2743

123

compared with microcellular PLA was 15%. Finally,

compared with the solid samples, the toughness of micro-

cellular samples was lower, possibly due to certain large

voids in the microcellular specimens due to the dynamic

nature of microcellular injection molding [54].

As with toughness, the strain-at-break of PLA (for both

solid and microcellular) increased for PLA–6%(H2004 ?

PA), PLA–12%(H2004 ? PA), and PLA–6%(H2004 ?

PA)–2%Nanoclay composite but decreased for PLA–

12%(H20 ? PA) (Fig. 10b). Also the strain-at-break val-

ues of the microcellular specimens were lower than those

of their solid counterparts.

The specific tensile modulus is shown in Fig. 10c.

Addition of HBP ? PA slightly decreased the specific

modulus. Among the solid specimens, addition of 6% and

12%(H2004 ? PA) decreased the specific modulus of solid

PLA by 3% and 10%, respectively. However, addition of

2% nanoclay did not have any effect on the specific

modulus of solid PLA–12%(H2004 ? PA). This may be

due to the fact that the nanoclays were not fully exfoliated

in the PLA matrix, as is evident from the TEM images

(Fig. 2) and XRD data (Fig. 1). Addition of 12%(H20 ?

PA) reduced the specific modulus by 11%. Thus, PLA–

12%(H20 ? PA) and PLA–12%(H2004 ? PA) had a

similar effect on the specific moduli of the specimens. In

general, the specific moduli of the microcellular specimens

showed a similar trend that of solid. However, the reduc-

tion in specific moduli of the microcellular PLA–HBP

blends was less compared with that of the solid samples.

For example, addition of 6%(H2004 ? PA) did not sig-

nificantly reduce the specific modulus of the microcellular

PLA, and addition of 12%(H2004 ? PA) reduced the

specific modulus by only 5%.

Figure 10d shows the specific tensile strength of all the

solid and microcellular specimens. The specific strength of

all samples was less than that of pure PLA (solid or micro-

cellular). Various strategies used to toughen a material often

results in a dramatic reduction in strength [39]. In this study,

the reduction in strength was observed to be 17–35% for

solid and 13–34% for microcellular specimens. In micro-

cellular specimens, lower strength reductions were found

when HBP was added to PLA compared to the solid speci-

mens. Compared with solid samples, microcellular samples

had lower average specific strengths. Again, this might be

caused by certain large voids in the microcellular specimens

due to the dynamic nature of the microcellular injection-

molding process [54]. Such large voids may concentrate

stress, thereby decreasing mechanical properties.

Fig. 10 Tensile properties of solid and microcellular PLA and PLA–HBP blends: a specific toughness, b strain-at-break, c specific modulus,

d specific tensile strength

2744 J Mater Sci (2010) 45:2732–2746

123

Conclusions

Solid and microcellular polylactide (PLA) and HBP blends

were injection-molded using conventional and microcel-

lular processes. For microcellular samples, a weight

reduction of 10–16% was achieved. The solid and micro-

cellular PLA–12%(H2004 ? PA)–2%Nanoclay compos-

ites revealed an intercalated structure based on the XRD

and TEM analyses. A quantitative analysis of the cell

morphology of the microcellular specimens showed that

the addition of HBPs and nanoclay decreased the average

cell size, and increased the cell density.

For all the solid and microcellular specimens, the degree

of crystallinity increased with the addition of HBPs ? PA

and nanoclay. Also, no significant difference was observed

between the degree of crystallinity of the corresponding

solid and microcellular samples. The storage moduli of both

the solid and microcellular samples decreased compared

with pure PLA. Among all the solid and microcellular

PLA–HBP blends, PLA–12%(H2004 ? PA)–2%Nano-

clay composites exhibited the highest specific toughness

(e.g., 405% increase for the solid specimen and 334%

increase for the microcellular specimen) and strain-at-break

(e.g., 626% increase for the solid specimen and 406%

increase for the microcellular specimen) followed by PLA–

12%(H2004 ? PA) and PLA–6%(H2004 ? PA). On the

other hand, PLA–12%(H20 ? PA) had a similar specific

toughness and strain-at-break values as the pure PLA for

both solid and microcellular samples. Furthermore, the

addition of HBPs ? PA and HBP–nanoclay caused a slight

reduction in specific modulus and a considerable reduction

in specific strength compared with pure PLA in all solid

and microcellular PLA–HBP blends. Finally, the micro-

cellular samples exhibited lower specific toughness, strain-

at-break, and specific strength compared with their solid

counterparts.

Acknowledgements We would like to acknowledge the financial

support from National Science Foundation (CMMI-0544729), the

USDA Forest Products Laboratory for the use of its equipment to

compound the materials and Perstorp Polyols Inc., USA for donating

the Boltorn HBPs.

References

1. Carole TM, Pellegrino J, Paster MD (2004) Appl Biochem Bio-

technol 113–116:871

2. Gross RA, Kalra B (2002) Science 297(5582):803

3. Kuriam JV (2005) In: Mohanty AK, Misra M, Drzal LT (eds)

Natural fibers, biopolymers, and biocomposites. CRC press, Boca

Raton

4. Mohanty AK, Misra M, Drzal LT (2002) J Polym Environ 10(1–

2):19

5. Pilla S, Gong S, O’Neill E, Yang L, Rowell RM (2009) J Appl

Polym Sci 111(1):37

6. Bhardwaj R, Mohanty AK (2007) J Biobased Mater Bioenergy

1:191

7. Heino A, Naukkarinen A, Kulju T, Tormala P, Pohjonen T,

Makela EA (1996) J Biomed Mater Res 30:187

8. Luciano RM, Zavaglia CAC, Duek EAR, Alberto-Rincon MC

(2003) J Mater Sci Mater Med 14:87

9. Itoh E, Matsuda S, Yamauchi K, Oka T, Iwata H, Yamaoka Y,

Ikada Y (2000) J Biomed Mater Res 53:640

10. Bleach NC, Nazhat SN, Tanner KE, Kellomaki M, Tormala P

(2002) Biomaterials 23:1579

11. Furukawa T, Matsusue Y, Yasunaga T, Shikinami Y, Okuno M,

Nakamura T (2000) Biomaterials 21:889

12. Park YJ, Nam KH, Ha SJ, Pai CM, Chung CP, Lee SJ (1997)

J Controlled Release 43:151

13. Giardino R, Fini M, Aldini NN, Giavaresi G, Rocca M (1999)

J Trauma 47:303

14. Lee SH, Kim BS, Kim SH, Kang SW, Kim YH (2004) Macromol

Biosci 4:802

15. Pego AP, Siebum B, Van Luyn MJA, Gallego Y Van Seijen XJ,

Poot AA, Grijpma DW, Feijen J (2003) Tissue Eng 9:981

16. Auras R, Harte B, Selke S (2004) Macromol Biosci 4:835

17. Grijpma DW, Zondervan GJ, Pennings AJ (1991) Polym Bull

25:327

18. Wehrenberg RH (1981) Mater Eng 94:63

19. Hiljanen-Vainio M, Karjalainen T, Seppala JV (1996) J Appl

Polym Sci 59:1281

20. Hiljanen-Vainio M, Orava PA, Seppala JV (1997) J Biomed

Mater Res 34:39

21. Buchholz B (1993) J Mater Sci Mater Med 4:381

22. Nakayama A, Kawasaki N, Arvanitoyannis I, Iyoda J, Yamamoto

N (1995) Polymer 36(6):1295

23. Joziasse CAP, Grablowitz H, Pennings AJ (2000) Macromol

Chem Phys 201:107

24. Kylma J, Seppaela JV (1997) Macromolecules 30(10):2876

25. Storey RF, Wiggins JS, Puckett AD (1994) J Polym Sci Part A

32(12):2345

26. Stolt M, Hiltunen K, Sodergard A (2001) Biomacromolecules

2(4):1243

27. Aslan S, Calandrelli L, Laurienzo P, Malinconico M, Migliaresi

C (2000) J Mater Sci Mater Med 35(7):1615

28. Hiljanen-Vainio M, Varpomaa P, Seppala J, Tormala P (1996)

Macromol Chem Phys 197(4):1503

29. Maglio G, Migliozzi A, Palumbo R, Immirzi B, Volpe MG

(1999) Macromol Rapid Commun 20(4):236

30. Maglio G, Malinconico M, Migliozzi A, Groeninckx G (2004)

Macromol Chem Phys 205(7):946

31. Meredith JC, Amis E (2000) Macromol Chem Phys 201(6):733

32. Kylma J, Hiljanen-Vainio M, Seppala J (2000) J Appl Polym Sci

76(7):1074

33. Kylma J, Seppala J (2000) J Appl Polym Sci 79(8):1531

34. Shibata M, Inoue Y, Miyoshi Y (2006) Polymer 47:3557

35. Hiljanen-Vainio M, Kylmae J, Hiltunen K, Seppaelae JV (1997)

J Appl Polym Sci 63(10):1335

36. Ljungberg N, Wesslen B (2005) Biomacromolecules 6:1789

37. Martin O, Averous L (2001) Polymer 42:6209

38. Bhardwaj R, Mohanty AK (2007) Biomacromolecules 8:2476

39. Lin Y, Zhang KY, Dong ZM, Dong LS, Li YS (2007) Macro-

molecules 40:6257

40. Wong S, Shanks RA, Hodzic A (2004) Macromol Mater Eng

289:447

41. Zhang W, Zhang Y, Chen Y (2008) Iran Polym J 17(12):891

42. Zhang JF, Sun X (2004) Polym Int 53:716

43. Jiang L, Wolcott MP, Zhang J (2006) Biomacromolecules 7:199

44. Seiler M (2002) Chem Eng Technol 25(3):237

45. Hong Y, Coombs SJ, Cooper-White JJ, Mackay ME, Hawker CJ,

Malmstrom E, Rehnberg N (2000) Polymer 41:7705

J Mater Sci (2010) 45:2732–2746 2745

123

46. Jannerfeldt G, Boogh L, Manson JAE (2000) Polymer 41:7627

47. Kil SB, Augros Y, Leterrier Y, Manson JAE (2003) Polym Eng

Sci 43(2):329

48. Mezzenga R, Boogh L, Manson JAE (2001) Compos Sci Technol

61(5):787

49. Okonishnikov GB, Blednykh EI, Skripov Mekh VP (1973)

Polimerov 2:370

50. Martini JE, Waldman FA, Suh NP (1982) In: SPE ANTEC

Technical Papers, 28: 674

51. Naguib HE, Park CB, Reichelt N (2004) J Appl Polym Sci

91:2661

52. Pilla S, Kramschuster A, Gong S, Chandra A, Turng LS (2007)

Int Polym Proc XXII(5):418–428

53. Kramschuster A, Pilla S, Gong S, Chandra A, Turng LS (2007)

Int Polym Proc XXII(5):436–445

54. Kramschuster A, Gong S, Turng LS, Li T, Li T (2007) J Biobased

Mater Bioenergy 1:37

55. Kramschuster A, Cavitt R, Ermer D, Chen Z, Turng LS (2005)

Polym Eng Sci 45(10):1408

56. Naguib HE, Park CB, Lee PC (2003) J Cell Plast 39(6):499

57. Pilla S, Kramschuster A, Lee J, Auer GK, Gong S, Turng LS

(2009) Compos Interfaces 16(7–9):869

58. Chandra A, Gong S, Yuan M, Turng LS (2005) Polym Eng Sci

45(1):52

59. Yuan M, Winardi A, Gong S, Turng LS (2005) Polym Eng Sci

45:773

60. Gong A, Turng L-S, Park CB, Liao L (2008) In: Mohanty AK,

Misra M, Nalwa HS (eds) Packaging nanotechnology. American

Scientific Publishers, USA

61. Kwag C, Manke CW, Gulari E (1999) J Polym Sci B 37(19):2771

62. Royer JR, Gay YJ, Desimone JM, Khan SA (2000) J Polym Sci B

38(23):3168

63. Kwag C, Manke CW, Gulari E (2001) Ind Eng Chem Res

40(14):3048

64. Suh NP (1996) In: Stevenson JF (ed) Innovation in polymer

processing-molding. Hanser Publishers, Munich

65. Gong S, Yuan M, Chandra A, Winardi A, Osorio A, Turng L-S

(2005) Int Polym Proc 2:202

66. Throne J (1979) In: Suh NP, Sung N (eds) Science and tech-

nology of polymer processing. MIT Press, Cambridge, MA, USA

67. Singh S, Ray SS (2007) J Nanosci Nanotechnol 7:2596

68. Pilla S, Gong S, Turng LS (2010) In: Mittal V (ed) Polymer

nanotube nanocomposites. Wiley-Scrivener, MA, USA

69. Rezgui F, Swistek M, Hiver JM, G’Sell C, Sadoun T (2005)

Polymer 46(18):7370

70. Garlotta D (2002) J Polym Environ 9:63

71. Nam JY, Ray SS, Okamoto M (2003) Macromolecules

36(19):7126

72. Wang H, Sun XZ, Seib P (2003) J Appl Polym Sci 90:3683

73. Naguib HE, Park CB, Reichelt N, Panzer U (2002) Polym Eng

Sci 42(7):1481

74. Lee LJ, Zheng C, Cao X, Han X, Shen J, Xu G (2005) Compos

Sci Technol 65(15–16):2344

75. Kharbas H, Nelson P, Yuan M, Gong S, Turng LS (2003) Polym

Compos 24(6):655

76. Hodge IM (1983) Macromolecules 16(6):898

77. Hodge IM, Huvard GS (1983) Macromolecules 16(3):371

78. Behrens AR, Hodge IM (1982) Macromolecules 15(3):756

79. Turi EA (1997) Thermal characterization of polymeric materials.

Academic Press, USA

80. Masirek R, Kulinski Z, Chionna D, Piorkowska E, Pracella M

(2007) J Appl Polym Sci 105(1):255

81. Pracella M, Chionna D, Anguillesi I, Kulinski Z, Piorkowska E

(2006) Compos Sci Technol 66(13):2218

82. Yasuniwa M, Tsubakihara S, Sugimoto Y, Nakafuku C (2004)

J Polym Sci B 42:25

83. Wang Y, Funari SS, Mano JF (2006) Macromol Chem Phys

207:1262

84. Pothan LA, Thomas S, Groeninckx G (2006) Compos A 37(9):

1260

85. Van Vlack LH (1989) Elements of materials science and engi-

neering. Addison-Wesley Publishing Company, USA

86. Yang L, Zhang C, Pilla S, Gong S (2008) Compos A 39:1653

2746 J Mater Sci (2010) 45:2732–2746

123